Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[joelkolb]

OS is 64 bit. The one that was working with 0.160 and 0.174 is Windows 10 (2004) and the files are on a mapped network share from an Unraid storage server. The one working with 0.225 is Windows Server 2019 and the files are on a locally attached Storage Spaces volume. Built in Windows Defender for A/V for both machines. No overclocking in either case. Both very different systems and storage but same type of results with clrmamepro so I don't think it is a system problem.

The settings you mention are intentional. I know it takes up a lot more space but fortunately I have lots of space. I really like having each game self contained without dependencies on any other files (mostly, except for CHDs and samples). So I do set clrmamepro for non-merged and do not separate BIOS files. Also, in some front ends like Emulation Station you can either set it to only show files in a precompiled XML list or to show all files in a particular folder. I like the flexibility of being able to quickly drop a ROM in the folder without having to edit the XML and have it show in the front end but I hate having BIOS files showing alongside the actual playable games for MAME. By not separating the BIOS files so that I don't need to have a separate BIOS file in the folder along with the games I can have things work just how I like them.

I'm glad that you were able to pinpoint the cause of this behavior with missing BIOS files and the stats and scan results. Hopefully you will be able to figure out how to fix it. I am happy to help. Thank you for working on this with me.

[Roman]

Well, I actually thought that your intention was to use not merged sets ;-)
...but there's a lack in your plan. Device sets. Nearly each set requires device sets, too...and they aren't put in a non merged sets and currently they are not handled like non-separated-biossets.
I already got requests to add something similar for device sets but had no time yet to add it.

For sure I will double check the missing tree print out very soon.

[Roman]

here we go: https://mamedev.emulab.it/clrmamepro/binaries/cmpro20201026a.7z

[joelkolb]

I tested this version and I'm a bit confused as to what it's doing. I used it to do a scan of my 0.160 set expecting that this time the stats would show 3 ROMs missing like before and the scan results would show me the 3 missing ROMs instead of nothing. But it actually fixed the set that was missing the ROMs. When I did a new scan everything came back clean with 0 missing and when I looked in the file hng64.zip it had all the files including the 3 that were missing before. Just to double check I made a backup of hng64.zip and deleted the 3 ROMs from the zip. I then did another full scan and it came back clean again and when I check the file the 3 ROMs were there. I'm not sure where it is getting these 3 ROMs from to fix the set. Also, shouldn't there be an indication in the stats that the set was fixed?

[Roman]

So what's confusing? fixing missing files works exactly like that.

hng64 missed 3 files and cmpro found the files in other places, so "fix missing" automatically added them and so hng64 won't appear in the scan results tree since it's ok now. If you don't turn on the fix options and you got the 3 files missing in hng64, you will see hng64 in the tree appearing. So it works as expected since the tree result only shows problems. The stats most likely shows something like 3/3 missed/fixed whatever.

cmpro looks in addpaths, backup folder and parent/clone relationship sets for missing files and in your case it found them somewhere there....

[joelkolb]

OK. I guess I was confused because it didn't fix the set in the previous version even though it was set to fix automatically. I was looking at it more like the problem was in the information it showed during and after the scan (stats and scan results) and not that it wasn't fixing files that it should have been fixing. That makes sense. If it fixes everything then there won't be any issues at the end. The stats did not show anything fixed though. It must have gotten the 3 missing files from clones because when I did the last scan with the copy of hng64.zip that I broke on purpose I also made sure the backup folder and add-paths were empty. Thanks!

[guynoirnpr]

I am getting the same results with my scans. I am using split sets (which I need to convert) but I get 20 missing ROMs but no listing of them. I understand you have looked at this problem already and it probably has to do with BIOS sets. I am using 0.226 MAME and cmp4039a_64 so everything is up to date there. One other problem is that the listing of "Missing Sets" now takes two days to complete when it used to take a minute.

Thank you for your update to cmp4039a_64 because I was having monumental problems with clrmamepro for a few months. But, cmp4039a_64 seems to have fixed all of my problems except for the BIOS sets. Thanks for all you do.

[Roman]

try https://mamedev.emulab.it/clrmamepro/binaries/cmpro64Test.7z

It creates a cmpro.log file in the cmpro main folder and lists all missing roms in there after a scan (and leaving cmpro). Would be interesting to see which are listed there. The file is locked until you quit cmpro and emptied on startup.


The code for the "missing sets" hasn't changed for ages and it's simply a directory walk/crawler over the files so it should be pretty fast. Depending on your amount of files and disk speed it should indeed only take up to a minute...but never 2 days. Sounds like some other app (maybe virusscanner) is interfearing....or something is locking/blocking access.

[guynoirnpr]

It turns out that the files were not bad but in another directory parallel to my directory called mame_bios_soft:
\
 |-mame_bios_soft               in ROM-Paths
 |-mame_bios_soft_bad        not in ROM-Paths
 |-mameroms                       in ROM-Paths
 |-mameroms_bad                not in ROM-Paths
Files were moved before the scan run from a correct directory (mame_bios_soft) to the parallel directory (mame_bios_soft_bad) that contained bad ROMs from previous scans. I keep bad old ROMs around for awhile just to make sure they are not needed. Of course those files had correct Checksums but names that had an extension added like crimep2_11.zip instead of crimep2.zip. This was done by clrmamepro in previous scans.
So now you know what the problem is when there are zero ROMs listed in the error list but ROMs showing as Missing. Certainly another weird thing that is still in clrmamepro.

[Roman]

Can you show me the cmpro.log output and not the missing list please.
There is a reason why  they are not listed in the scan results tree (but counted as missing). Actually I more expect the stats to be wrong than the output.

The miss list you've posted has most likely nothing to do with your missing stats. The "miss list" works totally different than the scan. As mentioned it is just a quick set lookup by file/folder name. It does nothing in detail at all. A relict from 20 years ago. Actually nobody should use that

The only important information is the scan results tree output. Even statistic count values are not really important. But of course it is interesting to see which files are counted as missing here....

[guynoirnpr]

cmpro.log is empty, even after ending cmpro. Yet it still shows files missing in the summary and nothing in the scan. Should I be using the version you provided in this thread?

[Roman]

Yes, you need to use https://mamedev.emulab.it/clrmamepro/binaries/cmpro64Test.7z
Simply replace the 64bit binary (keep a copy of the official one).

[guynoirnpr]

Yes I did finally use the test version and it had the same folders with the numerical extensions that came from the bad list (that took 2 days to produce), expanded to the individual files:

crimep2_11/cp2_1.dat
crimep2_11/cp2_2.dat
crimepat_12/cp_1.20_u1.bin
crimepat_12/cp_1.20_u2.bin
crimepat_14/cp01.dat
crimepat_14/cp02.dat
fastdraw_130/fast_01.bin
fastdraw_130/fast_02.bin
gallgall_21/gg_21_rom1.bin
gallgall_21/gg_21_rom2.bin
lastbh_006/bounty_01.bin
lastbh_006/bounty_02.bin
maddog2_100/md2_1.0_1.bin
maddog2_100/md2_1.0_2.bin
maddog2_110/md2_1.10_u1.bin
maddog2_110/md2_1.10_u2.bin
maddog2_202/md2_01.bin
maddog2_202/md2_02.bin
maddog_202/md_2.02_u1.bin
maddog_202/md_2.02_u2.bin
maddoga/maddog_01.dat
maddoga/maddog_02.dat
spacepir_14/sp_14_u1.bin
spacepir_14/sp_14_u2.bin
wsjr_15/wsjr151.bin
wsjr_15/wsjr152.bin
zortonbr_100/zb_u2.bin
zortonbr_100/zb_u3.bin

I have searched every directory that could have those files and found nothing. The versions of the files that I had around before, I moved to a completely different disk that is not shared with the computer that is running clrmamepro. So, I am at a loss as to where clrmamepro is finding those files. These files are all from alg_bios if you don't recognize them.

[Roman]

Did you check if you enabled all sets and all systems?
Maybe you got an option enabled that hides things. So please check the following and do a new full scan.

Scanner->Systems-> ensure that all checkboxes are checked (or hit all)
Scanner->Scan Results->Set Information->Ensure that all sets are selected (e.g. by hitting "select all")
Scanner->Scan Results->Context Menu ->View -> ensure that "hide fully missing sets" is NOT enabled
Scanner->Hash & CHD->Available Regions -> all checkboxes should be checked

If you like you can also send me your cmpro.ini file plus the belonging *.cmp file for your profile in cmpro's settings folder (I assume you're using an official MAME binary as data source)

[guynoirnpr]

Yes all my sets are selected. All systems are selected including alg_bios. All set information is selected. Hide fully missing sets is NOT enabled. All regions are selected.

Files attached

[Roman]

Thanks a lot for the files. I've found the problem for not listing the files. Very nice finding.
So update to 4.040

[guynoirnpr]

Glad you found it. I am a professional tester for HW/SW/FW etc. so I usually find things like this.

Additional:

When I ran 4.040 I had some problems. I did Rebuilder to add the missing files, then did a Scan. Mostly it was fine, but one thing that happened was that the first scan said the files were in where they should be in alg_bios but it showed that they were in mameroms also. I searched for the files in mameroms and could not see the files. I ran the scanner again and the second time it was correct with no files missing and no extra files in mameroms. I am very happy now. Thanks for 4.040 and all you do.

[Roman]

If you got dupes (identical sets exist in various rompaths) after running a rebuild, you may have rebuilder settings which don't match the scanner ones (e.g. you don't use sysdefpaths in rebuilder while you do in scanner). So a set might get rebuilt to a not wanted place accidently.
Anyway....Most likely the scan run cleaned them up. There are some rare cases where a second scan run can fix some stuff which a first one popped up.
Nothing really to worry...since your 2nd run seems to resolved it.